HIGHENERGY MONTE CARLO MODELS POTENTIAL IMPACT ON NEUTRON

HIGH-ENERGY MONTE CARLO MODELS – POTENTIAL IMPACT ON NEUTRON DOSIMETRY Vladimir Mares Helmholtz Zentrum München Institute of Radiation Protection 85764 Neuherberg Germany SATIF -13, HZDR Dresden, 10. - 12. October 2016

EURADOS EXERCISE V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Working Group WG-11: " High Energy Radiation Fields " Task 5: "Review of high-energy codes and models and their validation“ The exercise: calculate the Bonner sphereresponse § 3 He proportional counter SP 9 & 2 Bonner spheres: 9" PE, 9" PE(0. 5"Pb) § irradiation: parallel beam of mono-energetic neutrons from top § neutron energy: 1 Me. V, 100 Me. V, 1 Ge. V, 10 Ge. V § Monte Carlo codes: use different codes and reasonable INC models § calculate response and neutron fluence distributions inside of 3 He counter Bare 9“ PE W. Rühm, V. Mares, C. Pioch, S. Agosteo, A. Endo, M. Ferrarini, I. Rakhno, S. Rollet, D. Satoh, H. Vincke (2014) Comparison of Bonner sphere responses calculated by different Monte Carlo codes at energies between 1 Me. V and 1 Ge. V – Potential impact on neutron dosimetry at energies higher than 20 Me. V. Radiation Measurements 67, 24 -34 (2014). 9“ PE (0. 5“Pb)

EURADOS excercise Contributors: A. Endo, D. Satoh – JAEA, Japan S. Rollet – AIT, Austria S. Agosteo, M. Ferrarini – Polimi, Italy H. Vincke – CERN, Switzerland I. Rakhno – Fermilab, USA V. Mares, C. Pioch – HMGU, Germany Different Monte Carlo codes included: MCNPX, FLUKA, GEANT 4, PHITS, MARS Different models used: Bertini, Binary, ISABEL, INCL 4, CEM 03, JAM, JQMD, RAL, Abla, Dresner, GEM Cross Section Data ENDF/B-VI library; for 3 He ENDF/B-VII; the S(α, ) data for polyethylene; any high-energy library if available (LA 150, HILO 2 k, etc. ) Neutron fluence response: 1. Neutron fluence distributions inside of 3 He counter were folded in HMGU with the ENDF/B-VII cross section of the 3 He(n, p)3 H reaction < 100 Me. V 2. then normalised to HEMA 99 responses calculated by the MCNP and LAHET V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

9" PE (0. 5"Pb) sphere - anonymous comparison 10 Ge. V - higher than energy limit of many INC models used in excercise How can we trust the data? □ red squares refer to results obtained with GEANT 4 v. 8. 2 /Binary ∆ red triangles refer to results obtained with GEANT 4 v. 8. 2 /Bertini (Pioch et al. 2010) V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

HMGU – Extended-range Bonner Sphere Spectrometer (ERBSS) 18 measuring channels: § 3 He sph. detector SP 9 Centronic Ltd. § 15 PE-spheres from 2. 5” to 15” § 1 PE+Pb sphere (9” + 0. 5” Pb shell) § 1 PE+Pb sphere (9” + 1” Pb shell) § 1 SP 9 detector without PE sphere (bare) § detection of neutrons through 3 He(n, p)3 H (Q=764 ke. V, σth=5321 barns) § MSANDB unfolding code § initial guess spectra Bramblett, R. L. , Ewing, R. I. and Bonner, T. W. A new type of neutron spectrometer. Nucl. Instrum. Methods 9, 1– 12 (1960). V. Mares and H. Schraube, High energy neutron spectrometry with Bonner spheres, The IRPA Regional Symposium on Radiation Protection in Neighbouring Countries of Central Europe, Prague, Czech Republic, September 1997, 543 -547 V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Calculated and measured responses of Bonner spheres 9 -2“(Pb) Bare 3“ 6“ 7“ 9 -1“(Pb) 8“ 10“ 12“ 15“ • calculated with Monte Carlo codes: MCNP 4 A, HADRON, and LAHET (Mares et al. 1991, 1998) • monoenergetic neutrons at 12 energies between 1. 17 ke. V – 14. 8 Me. V in PTB, Braunschweig, Germany (Alevra et al. 1992) • thermal neutrons in NPL, Teddington, UK (Thomas et al. 1994) • high energy in RCNP, Osaka University, Japan (Mares et al. 2013) V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Impact on Unfolded Neutron Fluence & Dose Rates Neutron spectra measured at UFS Zugspitze (2650 m) MSANDB unfolding Different response functions used Differences in neutron fluence & H*(10) Thermal Epithermal Evaporation Cascade Total H*(10) ~ 4% ~ 14 % ~ 4% ~ 16 % ~ 9% C. Pioch, V. Mares, W. Rühm. Influence of Bonner sphere response functions above 20 Me. V on unfolded neutron spectra and doses. Radiation Measurements 45 (2010) 1263 -1267.

RCNP FACILITY OSAKA UNIVERSITY, JAPAN V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Research Center for Nuclear Physics - RCNP Osaka University Ring cyclotron facility Neutron experimental hall Ring Cyclotron (Up to 400 Me. V and 1 m. A for protons) 100 m tunnel for high-resolution neutron spectroscopy AVF Cyclotron (Up to 65 Me. V for protons) Beam dump V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Sketch of the experimental setup at 0° • target: 1 cm thick natural Li ( 6 Li 7. 6% and 7 Li 92. 4%) • 12 cm x 10 cm collimator (1. 5 m thick) with clearing magnet • proton beam intensity monitored by Faraday cup • BS measurements at 35 m distance from target V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

RCNP ring cyclotron facility, Osaka University Quasimono-energetic neutrons up to 387 Me. V Iron collimator Beam swinger magnet Neutron beam TOF tunnel 7 Li target V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

TOF MEASUREMENTS V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

TOF - Time-Of-Flight measurements SWINGER 100 m tunnel 25° 0° • energy spectra of 7 Li(p, n) neutron source measured with TOF method using organic scintillators NE 213 • 3 detectors of diameter: for neutron energy: measured at distance: 5. 08 cm 3 -10 Me. V 7 m 12. 7 cm 10 -100 Me. V 17 – 20 m 25. 4 cm ~ 100 Me. V 60 – 95 m • measurement down to 3. 5 Me. V at 0°, 5°, 10°, 15°, 20°, 25° • statistical uncertainties: below 3% • systematic uncertainty: 15 % mainly from detection efficiency of NE 213 (Y. Iwamoto et al. , 2013) V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Quasi-mono-energetic neutrons up to 387 Me. V 2014 0. 41 2011 2009 2014 2009 0. 44 Y. Iwamoto et al. , Nuclear Instruments and Methods in Physics Research A 804 (2015) 50– 58 V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Comparison of experimental and calculated results Y. Iwamoto, M. Hagiwara, D. Satoh, S. Araki, H. Yashima, T. Sato, A. Masuda, T. Matsumoto, N. Naka, T. Shima, T. Kin, Y. Watanabe, H. Iwase, T. Nakamura, Characterization of high-energy quasi-monoenergetic neutron energy spectra and ambient dose equivalents of 80– 389 Me. V 7 Li(p, n) reactions using a time-of-flight method, Nuclear Instruments and Methods in Physics Research A 804 (2015) 50– 58 V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

NEUTRON SPECTROMETRY RESULTS V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Neutron fluence spectra – ERBSS and TOF results 100 Me. V and 296 Me. V protons 7 Li(p, n) - at 35 m from target and 0° • unfolded HMGU-ERBSS neutron spectra • TOF spectra are re-binned to ten bins per decade • HMGU-BSS spectra are very close to TOF spectra above 10 Me. V • evaporation peak around 2 Me. V only in HMGU-BSS spectra V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Neutron ambient dose equivalent, H*(10), in Sv/C at 35 m from target * Contribution above 3 Me. V Partial contribution to the total neutron ambient dose equivalent, H*(10), in % V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Comparison of Bonner sphere responses calculated by different MC H*(10) V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Comparison of Bonner sphere responses calculated by different MC H*(10) V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

BSS calibration in terms of response values • measured count rates were corrected by subtraction of count rates below 3. 5 Me. V • remaining count rates were divided by TOF integral fluence • effective energy were estimated for each sphere using its response and TOF fluence rate distribution October 2011 November 2009 V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

PARTICLE(HADRON) THERAPY V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Particle (Hadron) Therapy CCB-Krakow gantry room, Poland. Photo: V. Mares, 2014 HIT horizontal beam, Heidelberg, Germany. Photo: V. Mares, 2015 External beam radiotherapy is using beams of energetic protons, neutrons, or positive ions for cancer treatment. Most common is proton therapy. Proton therapy is advanced clinical modality – at present about 50 centres in the world Around 100, 000 patients worldwide treated with protons since 1990 (Loma Linda, CA, the first hospital-based treatment) More than 10, 000 treated with heavier ions, generally carbon Various company offer turn-key solutions for medical centers (IBA, Varian, …) V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Particle (Hadron) Therapy Need for improved high-energy neutron dosimetry The basic problem: the “radiation bath” Sources of out-of-field doses: scattering in nozzle, patient, and air Organ at risk (OAR) Secondary neutrons may increase the risk of a second cancer significantly Objective To assess non-target organ doses in radiotherapy and the related risks of second cancers The emphasis on dosimetry methods ~ 10 m. Gy ~ 60 Gy (~ 1 : 6000) R. M. Harrison, Introduction to dosimetry and risk estimation of second cancer induction following radiotherapy, Radiation Measurements 57 (2013) 1 -8 V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Particle (Hadron) Therapy March 2014: 44 proton centers 7 heavy ion centers Under construction: 25 proton centres 4 heavy ion centers V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Spectrometryof Stray Neutron Radiation Field in CCB-Kraków PTC Neutron spectra measured around the ten years-old CIRS phantom using the UAB and HMGU ERBSS system in 2014. V. Mares, M. Romero-Expósito, J. Farah, S. Trinkl, C. Domingo, M. Dommert, L. Stolarczyk, L. Van Ryckeghem, M. Wielunski, P. Olko, R. M. Harrison, A comprehensive spectrometry study of stray neutron radiation field in scanning proton therapy, Phys. Med. Biol. 61 (2016)

Spectrometryof Stray Neutron Radiation Field in CCB-Kraków PTC Relative contributions: thm: thermal (1 me. V E < 0: 4 e. V), int: intermediate (0: 4 e. V E < 100 ke. V), fst: fast (100 ke. V E <19: 6 Me. V) and hgh: high (E 19: 6 Me. V) The effect of different response functions is somewhat hidden in unfolding process M. Dommert, Master Thesis, TUM 2015 V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Spectrometryof Stray Neutron Radiation Field in CCB-Kraków PTC CIRS ATOM® Phantoms M. Dommert, Master Thesis, TUM 2015 GEANT 4 v. 10. 0. simulated guess spectra for the MSANDB unfolding code; the Bertini INC model (black) and the Binary INC model (red): Voxel phantom, simplified beam model, without treatment room V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

EURADOS INITIATIVE V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

EURADOS Initiative: Task 10: “Benchmarking” for high energy. MC models– potential impact on dosimetry (V. Mares) • • New task created in Milan AM 2016 as common activity among three WG’s: • WG 11 - High energy radiation fields (J-F. Bottollier Depois) • WG 9 - Radiation dosimetry in radiotherapy (R. Harrison) • WG 6 - Computational dosimetry (R. Tanner) The first tasks: • to find an appropriate task title; what energy range? • to invite experts (nuclear data, evaluation and validation, high-energy models, etc. ) • to define the objectives and methods, clarify actions, outline a roadmap • POTENTIAL BENCHMARK EXERCISES V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Next EURADOS ANNUAL MEETING will take place on Feb 27 th - Mar 2 nd 2017 at KIT in Karlsruhe, Germany V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

CONCLUSIONS • HMGU-ERBSS (HEMA 99) and TOF neutron spectra at 0° are very close above ~5 Me. V • TOF H*(10) are systematically lower because of energy threshold ~3 Me. V • contribution of evaporation neutrons to the total H*(10) is ~10 % • contribution of peak neutrons to the total H*(10) is ~40 % • use of 3 different response matrices (MCNP/LAHET, GEANT 4 -Bertini, GEANT 4 -Binary) • all three calculated response matrices are almost equally consistent with measurements • uncertainties in high-energy models very important in hadron therapy – treatment planning • EURADOS: new task created dealing with uncertainties in cross sections and highenergy models and their impact on neutron dosimetry – you are very welcome to join us send me e-mail please: mares@helmholtz-muenchen. de V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

THANKS FOR YOUR ATTENTION V. Mares, SATIF -13, HZDR Dresden, 10. - 12. October 2016

Need for improved high-energy neutron dosimetry Fluence-to-H*(10) conversion coefficients from ICRP 1997 extended to high energies with data from Pelliccioni 2000, spline by Mares FAST ENERGY HIGH ENERGY e. V M M 0 10 300 V. Mares, ICRS-13&RPSD-2016 October 3|6 Paris

HEMA 99 - BSS Response functions • 1991 - calculated with the MCNP 3 B 3 for 10 me. V < E <30 Me. V • 1997 - extended to 1. 5 Ge. V by the MC codes MCNP 4 A, HADRON, and LAHET • 1998 - MCNP 4 A and LAHET calculation for 9” spheres with lead, extended to 10 Ge. V • 1999 - response matrix with 130 x 24 grid __________________________________ • neutron x-section data of H extracted from ENDF/B-IV Rev. 1, 3 He from ENDF/B-III, and C from LASL-SUB 1976 • MCNP 4 A code extended up to 100 Me. V using the n-x-sections for H and C from the LA 100 • thermal S( , β) tables for PE from a special ENDF • in LAHET: Bertini or ISABEL cascade model, Fermi break-up model, multistage pre-equilibrium model • in HADRON: cascade model of Dubna, exciton model for pre-equilibrium stage MC calculated and measured Φ-response (Mares et al. 1991, 1998; Alevra et al. 1992, Thomas et al. 1994)

Calculated and measured responses of Bonner spheres • calculated with Monte Carlo codes: MCNP 4 A, HADRON, and LAHET (Mares et al. 1991, 1998) • monoenergetic neutrons at 12 energies between 1. 17 ke. V – 14. 8 Me. V in PTB, Braunschweig, Germany (Alevra et al. 1992) • thermal neutrons in NPL, Teddington, UK (Thomas et al. 1994) Bare 3. 5“ 3“ HEMA 99 4“ 4. 5“ 9 -2“(Pb) 5“ 6“ 9 -1“(Pb) 7“ 8“ 10“ 12“ 15“